========= Content: --------- <doc id='1'>量化周报 科创 50 即将确认日线下跌，风格切换正在进行 科创 50 即将确认日线下跌，风格切换正在进行。本周（11.10-11.14）， 大盘横盘震荡，上证指数全周收跌 0.18%。在此背景下，汽车确认日线级 别下跌，农林牧渔、消费者服务迎来日线级别上涨。市场的本轮上涨自 4 月 7 日以来，日线级别反弹已经持续了 7 个多月，反弹幅度也基本在 30% 左右，各大指数和板块的上涨基本都轮动了一遍，超 2/3 的行业日线级别 上涨处于超涨状态，几乎所有的规模指数及一半以上的行业更是走出了复 杂的 9-17 浪的上涨结构，而食品饮料、医药、商贸零售、汽车也已经形成 了日线级别下跌，军工、传媒也有较大概率将确认日线级别下跌。因此我 们认为本轮日线级别上涨大概率已临近尾声。此外，创业板、科创 50 自 6 月份以来的上涨短期内已基本见到顶部，科创 50 即将迎来日线级别下 跌，未来科技板块大概率会是震荡调整的态势，风格切换正在进行。中期 来看，上证指数、上证 50、沪深 300、中证 500、深证成指、创业板指、 科创 50 纷纷确认周线级别上涨，而且在日线上只走出了 3 浪结构，中期 牛市刚刚开始；此外，已有 27 个行业处于周线级别上涨中，且 19 个行业 周线上涨走了 1-3 浪结构，因此我们认为本轮牛市是个普涨格局。中期对 于投资者而言，仍然可以逆势布局。</doc> <doc id='2'>A 股景气指数观察。截至 2025 年 11 月 14 日，A 股景气指数为 20.94， 相比 2023 年底上升 15.51，当前处于上升周期中。 A 股情绪指数观察。当前 A 股情绪见底指数信号：空，A 股情绪见顶指数 信号：空，综合信号为：空。 指数增强组合本周表现尚可。中证 500 增强组合跑赢基准 0.66%，沪深 300 增强组合跑输基准 0.58%。 风格上，当前残差波动率因子占优。从纯因子收益来看，本周保险、医药、 有色金属等行业因子相对市场市值加权组合跑出较高超额收益，计算机、 汽车等行业因子回撤较多；风格因子中，残差波动率因子超额收益较高， 市值呈较为显著的负向超额收益。从近期因子表现来看，高杠杆股表现优 异，市值、成长等因子表现不佳。 风险提示：量化周报观点全部基于历史统计与量化模型，存在历史规律与 量化模型失效的风险。</doc> <doc id='3'>作者 分析师 刘富兵 执业证书编号：S0680518030007 邮箱：liufubing@gszq.com 分析师 沈芷琦 执业证书编号：S0680521120005 邮箱：shenzhiqi@gszq.com 分析师 张国安 执业证书编号：S0680524060003 邮箱：zhangguoan@gszq.com 分析师 赵博文 执业证书编号：S0680524070004 邮箱：zhaobowen@gszq.com 分析师 汪宜生 执业证书编号：S0680525070003 邮箱：wangyisheng@gszq.com 研究助理 阮俊烨 执业证书编号：S0680124070019 邮箱：ruanjunye@gszq.com 相关研究 1、《量化分析报告：择时雷达六面图：本周基本面改善， 拥挤度下降》 2025-11-08 2、《量化分析报告：六周期框架下的多资产 ETF 配置》 2025-11-05 3、《量化点评报告：十一月配置建议：关注小盘+价值 的均衡配置》 2025-11-03</doc> <doc id='4'> | 1. 市场走势分析 4 | | | --- | --- | | 1.1 科创 即将迎来日线下跌，风格切换正在进行 50 4 | | | 1.2 创业板短期或已基本见顶 5 | | | 2. 市场行业分析 6 | | | 2.1 中期看多煤炭、房地产、石油石化、消费者服务 6 | | | 2.2 汽车确认日线下跌，农林牧渔、消费者服务迎来日线上涨 7 | | | 3. 市场景气与情绪观察 8 | | | 3.1 A 8 | 股景气指数观察 | | 3.2 A 股情绪指数观察 9 | | | 4. 情绪指标择时和主题投资机会 11 | | | 4.1 半导体概念股机会 11 | | | 4.2 中证 500 增强组合 11 | | | 4.3 沪深 300 增强组合 14 | | | 5. 市场风格分析 15 | | | 5.1 风格因子表现 15 | | | 5.2 市场主要指数收益风格归因 17 | | | 风险提示 18 | | | 附录： 19 | | | 指数分析及投资建议 19 | | | 行业分析及投资建议 19 | |</doc> <doc id='5'> | 图表 | 1： | 上证综指走势结构图 | | 4 | | --- | --- | --- | --- | --- | | 图表 | 2： | 上证 50 | 走势结构图 | 4 | | 图表 | 3： | 沪深 300 | 走势结构图 | 5 | | 图表 | 4： | 中证 500 | 走势结构图 | 5 | | 图表 | 5： | 中证 1000 | 走势结构图 | 5 | | 图表 | 6： | 深证成指走势结构图 | | 5 | | 图表 | 7： | 创业板指走势结构图 | | 6 | | 图表 | 8： | | 消费者服务中期走势结构图 | 6 | | 图表 | 9： | 煤炭中期走势结构图 | | 6 | | 图表 | 10： | | 石油石化中期走势结构图 | 7 | | 图表 | 11： | | 房地产中期走势结构图 | 7 | | 图表 | 12： | | 当前下行周期与历史平均趋势（横坐标： 持续天数） | 8 | | 图表 | 13： | A 股景气度指数 | | 9 | | 图表 | 14： | | 波动-成交情绪时钟收益统计（沪深 300） | 9 | | 图表 | 15： | A | 股情绪指数（见底预警指数） 10 | | | 图表 | 16： | A | 股情绪指数（见顶预警指数） 10 | | | 图表 | 17： | A | 股情绪指数系统择时表现 11 | | | 图表 | 18： | 半导体概念股 | 11 | | | 图表 | 19： | 中证 500 | 增强组合表现 12 | | | 图表 | 20： | 中证 500 | 增强组合持仓明细 12 | | | 图表 | 21： | 沪深 300 | 增强组合表现 14 | | | 图表 | 22： | 沪深 300 | 增强组合持仓明细 14 | | | 图表 | 23： | | 近一周十大类风格因子暴露相关性 16 | | | 图表 | 24： | | 近一周十大类风格纯因子收益率 16 | | | 图表 | 25： | | 近一周行业纯因子收益率 16 | | | 图

Core Insights - The article discusses the advancements in quantitative research by Galaxy Fund, focusing on the integration of linear and nonlinear analysis to enhance stock selection and uncover investment opportunities [1][2]. Group 1: Quantitative Research Approach - Galaxy Fund's quantitative team has shifted from traditional linear analysis to nonlinear methods to better adapt to complex market environments [1]. - The combination of multi-factor models and nonlinear machine learning models aims to achieve stable excess returns while reducing tracking error [1][2]. Group 2: Strategy Development - The quantitative strategies include both linear strategies, primarily using common multi-factor models, and nonlinear strategies such as XGBoost and LightGBM [2]. - XGBoost is highlighted for its ability to rank factor importance, enhancing the model's adaptability to market changes, especially in structured market conditions [2]. Group 3: Neural Network Learning - The development of complex neural network learning is emphasized, where long-term rules and short-term information are combined to improve model training [3]. - This approach helps in quickly adapting the quantitative model to market fluctuations by refining the feature extraction process [3]. Group 4: Satellite Strategies - In addition to the main strategies, satellite strategies such as dividend selection and large-cap growth selection are employed to further enhance market adaptability [4]. - The dividend selection strategy focuses on high dividend yield stocks, while the large-cap growth strategy targets large-cap, high-growth stocks [4]. Group 5: Product Offerings - Galaxy Fund has launched two index enhancement products: the Galaxy CSI 300 Index Enhanced Fund and the Galaxy CSI A500 Index Enhanced Fund, with plans for a Galaxy CSI 800 Index Enhanced Fund [4][5]. - The CSI 800 Index is noted for its balanced representation of both large-cap blue-chip and mid-cap growth styles, covering a wide range of sectors in the Chinese economy [5].

Core Insights - The rapid development of AI technology is significantly enhancing the power of quantitative investment, leading to increased market attention on public quantitative investment strategies [1] - Huang Zhigang, Assistant General Manager and Head of Quantitative Investment at Nanhua Fund, emphasizes the limitations of traditional multi-factor models, which are based on historical data and fail to address long-term market effectiveness [1][4] - Huang identifies three key issues that an excellent quantitative investment model must solve: constructing investment safety margins, identifying value traps, and reasonably defining company prices [4][5] Quantitative Investment Framework - Huang's quantitative investment framework is summarized as "value stock selection and dual rotation," focusing on core factors such as Dividend Payout Ratio (DR), Return on Equity (ROE), and Earnings Yield (EP) [2] - The first step in the model involves predicting each company's ROE and EP, followed by calculating the Potential Return (IR) and ranking stocks based on IR values to build an investment portfolio [2][4] - The approach aims to find good companies at good prices, leveraging the objectivity, efficiency, and discipline of quantitative investment [2][5] Stock Selection and Adjustment - Stocks selected through this method are not static; they are continuously adjusted based on factor changes [3] - Huang constructs a foundational stock pool by selecting stocks that have declined significantly over the past 3 to 5 years, updating this pool daily to achieve a "buy low, sell high" strategy [3] Performance Metrics - As of now, Huang manages four funds with a total scale exceeding 1 billion yuan, with notable performance metrics such as a net value growth rate of over 87% for Nanhua Fenghui Mixed A since inception [4] - The Nanhua Fengyuan Quantitative Stock Selection Mixed A, managed since January 2024, has achieved a net value growth rate exceeding 38% [4] Risk Management and Strategy - Huang highlights the importance of balancing "good companies" and "good prices," aiming for a better equilibrium between the two rather than focusing solely on short-term performance [5] - The quantitative investment strategy includes risk control measures such as maintaining a diversified portfolio, limiting individual stock weight, and ensuring a balanced strategy style [8] - The fund's turnover rate is kept stable at around 12 times, with a holding range of 80 to 130 stocks, aiming to smooth out volatility risks through relative excess returns [7][8]

Core Insights - The article discusses the advancements in quantitative research by Galaxy Fund, focusing on the integration of linear and nonlinear analysis to enhance stock selection and investment opportunities [1][2] Group 1: Quantitative Research Strategies - Galaxy Fund's quantitative team has shifted from traditional linear analysis to nonlinear methods to better adapt to market changes and identify investment opportunities [1] - The combination of multi-factor models and nonlinear machine learning models aims to achieve stable excess returns while reducing tracking error in the overall portfolio [1][3] - The team has developed strategies that include both linear approaches, primarily using multi-factor models, and nonlinear methods such as XGBoost and LightGBM [1][2] Group 2: Neural Network Development - The team is advancing from simple neural network learning to more complex neural network models to improve market adaptability [2] - By integrating long-term rules with short-term information, the team enhances the feature extraction process, allowing for better training of supervised neural networks [2] Group 3: Satellite Strategies - In addition to the main strategies, satellite strategies such as dividend selection and large-cap growth selection are employed to further enhance market adaptability [3] - The dividend selection strategy focuses on high dividend yield stocks, while the large-cap growth strategy targets stocks with high market capitalization and growth potential [3] Group 4: Product Offerings - Galaxy Fund has launched two index enhancement products: the Galaxy CSI 300 Index Enhanced Fund and the Galaxy CSI A500 Index Enhanced Fund [3] - A new product, the Galaxy CSI 800 Index Enhanced Fund, is in the process of being issued, which aims to provide a balanced representation of both large-cap blue-chip and mid-cap growth styles [4] Group 5: Index Characteristics - The CSI 800 Index is noted for its balanced coverage of A-share assets, representing the overall vitality of the Chinese economy across various sectors [4] - The index encompasses 31 primary industry categories, including traditional sectors like banking and emerging sectors such as electronics and pharmaceuticals [4]

Core Insights - The rapid development of AI technology is significantly enhancing the power of quantitative investment, leading to increased market attention on public quantitative investment strategies [1] - Huang Zhigang, Assistant General Manager and Head of Quantitative Investment at Nanhua Fund, emphasizes the limitations of traditional multi-factor models in quantitative investment, which are primarily based on historical data and fail to address long-term market effectiveness [1][4] - Huang identifies three critical issues that an excellent quantitative investment model must solve: constructing investment safety margins, identifying value traps, and reasonably defining company prices [4][5] Group 1: Investment Strategy - Huang's quantitative investment framework is summarized as "value stock selection and dual rotation," focusing on core factors such as Dividend Payout Ratio (DR), Return on Equity (ROE), and Earnings Yield (EP) [2] - The first step in the quantitative model involves predicting each company's ROE and EP, followed by calculating the Potential Return (IR) and ranking stocks based on IR values to build an investment portfolio [2][4] - The model aims to find good companies at good prices, leveraging the objectivity, efficiency, and discipline of quantitative investment over subjective human judgment [2][4] Group 2: Stock Selection and Risk Management - Stocks selected through this method are not static; they are continuously adjusted based on changing factors, with a focus on building a stock pool from those that have declined significantly over the past 3 to 5 years [3] - Huang employs a dual approach to avoid value traps and select stocks with low price-to-earnings ratios, low price-to-book ratios, and high dividend yields to provide safety margins [5][7] - The investment goal is to balance between "good companies" and "good prices," seeking long-term performance advantages rather than focusing on short-term results [5] Group 3: Performance and Market Position - As of now, Huang manages four funds with a total scale exceeding 1 billion yuan, with notable performance metrics such as a net value growth rate of over 87% for Nanhua Fenghui Mixed A since inception [4] - Huang acknowledges the increasing competition in the quantitative investment space, which makes it more challenging to obtain alpha, necessitating continuous updates and factor exploration in quantitative models [4][7] - The domestic quantitative investment market is still developing compared to mature foreign markets, with public quantitative investment expected to gradually reveal its advantages in fundamental research [7]

Group 1 - The report highlights the performance of the "Xinxuan ETF Absolute Return Strategy," which achieved an annualized return of 14.23% over the past three years, with a maximum drawdown of only 8.6% and a Sharpe ratio of 1.44 [10] - As of 2024, the total return of the Xinxuan ETF portfolio is 54.04%, outperforming the equal-weighted ETF by 11.1%, with a Sharpe ratio of 1.55 and a maximum drawdown of 6.3% [10] - The latest holdings of the Xinxuan ETF strategy include various ETFs such as the Innovation Drug ETF (15%) and the Bank ETF (10%) [11] Group 2 - The "All-Weather Multi-Asset Risk Parity Strategy" has yielded a return of 27.75% since the beginning of 2024, with a maximum drawdown of 3.62% and a Sharpe ratio of 2.56 [13] - This strategy diversifies across different assets and strategies, including gold ETFs and U.S. equity ETFs, to enhance returns while reducing overall portfolio volatility [15] Group 3 - The "Recovery Fixed Income+" strategy aims to balance inflation and credit factors while maintaining liquidity, utilizing a monthly rotation among 15 high-liquidity ETFs in the Hong Kong market [19] - Since 2021, this strategy has achieved an annualized return of 7.63% with a volatility of 7.06% and a Sharpe ratio of 1.07 [19] Group 4 - The "China-U.S. Core Asset Portfolio" includes strong trend assets such as liquor, dividends, gold, and the Nasdaq, achieving an annualized return of 33.66% since early 2015, outperforming equal-weighted indices by 12.11% [21] - The latest holdings in this portfolio include the Dividend ETF [23] Group 5 - The "High Prosperity/Dividend Rotation Strategy" has generated an annualized return of 25.49% since early 2021, significantly outperforming equal-weighted indices by 22.91% [26] - The strategy adjusts holdings based on economic signals, switching between high-growth ETFs and dividend ETFs [26] Group 6 - The "Double Bond LOF Enhancement Strategy" has achieved an annualized return of 6.43% since early 2019, with a Sharpe ratio of 2.48 and a maximum drawdown of 2.42% [29] - This strategy focuses on increasing the weight of bonds in the portfolio while maintaining exposure to other assets [29] Group 7 - The "Structured Risk Parity Strategy (QDII)" has yielded a return of 28.53% since the beginning of 2024, with a maximum drawdown of 2.38% and a Sharpe ratio of 2.57 [32] - This strategy combines domestic long-term bond ETFs with QDII equity products and gold to enhance returns [32] Group 8 - The report indicates that 24 new public funds were established this week, raising a total of 141.73 billion yuan, with 14 new index funds accounting for 65.90 billion yuan of this total [39] - The new index funds include various themes such as technology, agriculture, and energy [39] Group 9 - As of November 14, 2025, A-share, bond, commodity, and cross-border ETFs saw net subscription amounts of 122.0 billion yuan, -2.7 billion yuan, 59.4 billion yuan, and 102.4 billion yuan, respectively [49] - In the A-share ETF segment, the net inflow was led by sectors such as electric power equipment and new energy [50]

- The report constructs an ETF hotspot trend strategy based on the highest and lowest price trends of ETFs, selecting those with both highest and lowest prices in an upward trend. Further, it constructs a support-resistance factor based on the relative steepness of the regression coefficients of the highest and lowest prices over the past 20 days, and selects the top 10 ETFs with the highest turnover rate in the past 5 days/20 days to construct a risk parity portfolio[27][30] - The report tracks the performance of various style factors, noting that the value factor recorded a positive return of 2.36%, the leverage factor recorded a positive return of 1.08%, and the volatility factor slightly rebounded with a return of 0.19%[41][42] - The report evaluates the performance of different alpha factors, highlighting that the quick ratio factor had the best performance with a weekly excess return of 1.32%, followed by the debt-asset ratio factor with a weekly excess return of 1.21%, and the earnings variability over 5 years factor with a weekly excess return of 1.04%[44][46][47] - The ETF hotspot trend strategy recorded a cumulative excess return over the CSI 300 index since the beginning of the year[28][29] - The value factor achieved a weekly return of 2.36%, the leverage factor achieved a weekly return of 1.08%, and the volatility factor achieved a weekly return of 0.19%[41][42] - The quick ratio factor achieved a weekly excess return of 1.32%, the debt-asset ratio factor achieved a weekly excess return of 1.21%, and the earnings variability over 5 years factor achieved a weekly excess return of 1.04%[44][46][47]

- The report does not contain specific quantitative models or factors for analysis [1][2][3] - The report primarily focuses on market performance, style indices, valuation metrics, and fund flows without detailing quantitative models or factor construction [1][2][3] - Key metrics such as PE_TTM, ERP, and style index performance are discussed, but no explicit quantitative model or factor development process is provided [41][51][59]

- The report discusses the microstructure rebalancing in the A-share market, highlighting the increased concentration of stock price movements driven by speculative capital inflows since June 2025, which has impacted quantitative products' portfolio construction and risk exposure adjustments[13][23][24] - Quantitative products have adjusted their exposure to micro-cap stocks, initially reducing their holdings to mitigate nonlinear market cap risks, and later increasing allocations to amplify excess returns as speculative capital inflows weakened post-October 2025[13][23][24] - The report emphasizes the Barra style factor performance, noting that fundamental factors such as BP value and investment quality have shown positive returns, while transaction-related factors like short-term momentum have also delivered strong excess returns during the market's recent fluctuations[23][24][25]

- The report does not contain any specific quantitative models or factors for analysis or construction[1][2][3][4] - The report primarily focuses on market trends, fund flows, and industry performance without detailing quantitative models or factors[5][6][7] - No formulas, construction processes, or evaluations of quantitative models or factors are provided in the report[8][9][10]